This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. Primary support for the subproject and the subproject's principal investigator may have been provided by other sources, including other NIH sources. The Total Cost listed for the subproject likely represents the estimated amount of Center infrastructure utilized by the subproject, not direct funding provided by the NCRR grant to the subproject or subproject staff. Accurate RNA recognition by other biomolecules such as proteins, cofactors and other RNA molecules are critical to many cellular functions. Employing a variety of computational chemistry tools such as molecular dynamics simulations, quantum calculations and hybrid quantum mechanical/molecular mechanics methods, our research examines three primary areas, messenger RNA - transfer RNA (mRNA-tRNA) recognition, a key step in the translation of proteins, protein-RNA interactions in human immunodeficiency virus (HIV) and pKa calculations in catalytic RNA molecules. In the first area, the role of naturally occurring, posttranscriptionally modified bases in affecting tRNA-mRNA recognition is examined. In human tRNALys,3, we have found that a modified base at position 37 are required for maintenance of a canonical stair-stepped conformation in the anticodon bases (34-36). Ab initio studies employing natural bond orbital analysis with the M05-2X functional are underway to determine the underlying stabilizing forces and the role of modified bases at the 37th position in retaining a stair-stepped conformation in all tRNAs. Optimization of hydrogen positions at the M05-2X/6-31+G(d,p) theory level needs to be carried out for tetranucleotides and trinucleotides (dimers are ~1400 basis functions), which on our local machines can take greater than 45 days/calculation. Faster computing resources are required to make progress on this project. In the second area of research, we are examining the role of water and electrostatics in RNA-peptide recognition. In late phase Rev-RRE recognition mediates nucleocytoplasmic export of partially and unspliced HIV mRNA. From in vitro selection studies performed by Frankel and coworkers, a synthetic peptide known as RSG-1.2 has been found to bind RRE with greater affinity and specificity than the native Rev peptide. We have simulated both Rev and RSG-1.2 peptides complexed with the RRE RNA in explicit water using AMBER and have found a correlation between water structure in the peptide-RNA complexes and binding affinity. More simulations to corroborate earlier findings are required. Systems are roughly 35,000 atoms and data could be collected more efficiently employing parallel AMBER code. Lastly, in collaboration with Darrin York, we are calculating pKas in catalytic RNA molecules known as ribozymes. The thermodynamic integration methods require equilibrated starting systems. Current systems are carried out in explicit solvent (TIP4Pew), include 150 mM NaCl buffer solution beyond the neutralized RNA and are about 75,000 atoms. These systems require a number of simulated annealing rounds to equilibrate the ion atmosphere and then the RNA must be subsequently equilibrated in the presence of the buffer before TI calculations can be performed. This allocation is requested to take advantage of parallel computing facilities while also exploring optimum Teragrid platforms for future allocation requests.